11.

INTEGRATING THE FOUR SKILLS: CURRENT AND HISTORICAL

PERSPECTIVES

Eli Hinkel
In the contemporary world of second and foreign language teaching, most professionals
largely take it for granted that language instruction is naturally divided into discrete skill sets,
typically reflecting speaking, listening, reading, and writing, and usually arranged in this order.
That is, the primacy of speaking skills has remained unquestioned, at least in North America, for
almost the entire past century, since the rise and preeminence of structural linguistics in second
and foreign language teaching (Language Teaching, 2007). Based on the principles of
Bloomfieldian linguistic analyses and their applications to language pedagogy, the structural
division of language teaching in the four skill areas has the learning objective of imitating the
native speaker. The continual separation of the four skills lies at the core of research and testing
in speaking, listening, reading, and writing.
Some current approaches to teaching language, however, strive to integrate the four skills
in pedagogy whenever possible. Integrated language teaching and various integrated
pedagogical paradigms are usually associated with outgrowths of communicative teaching.
Relative to its predecessor, the audiolingual method, integrated teaching of the four skills
represents a central innovation. On the other hand, in the U.K., the path toward integrated
teaching of the language skills did not derive from a strong audiolingual focus but rather from an
evolution of older situational and functional teaching methods that were developed prior to and
concurrent with the structural method in the U.S. Current models of integrated language
teaching are not without their shortfalls. Nor is integrated instruction appropriate in all contexts
of language teaching and for all purposes of language learning. The advantages and
disadvantages of integrated teaching may crucially determine its usefulness in second or foreign
language contexts.
This chapter begins with a brief look at the historic and methodological reasons for the
continual separation of the four skills in teaching. Modern-day perspectives on skill integration
and integrated curriculum designs will also be discussed, together with problems and issues
typically associated with integrated teaching. The chapter will then address the highly
idiosyncratic and limited designs of major English language tests in the U. K. and the U. S., as
well as the indelible effect of tests on the separation of the four skills. The chapter concludes
with an overview of the pedagogical and methodological currents in integrated language
instruction.

How the Four Skills Became Separated
Contemporary methods for teaching second and foreign languages in the U. S. and the U.
K. have followed two distinctly different routes, primarily due to the divergent histories of the
two countries during the second half of the 20th century. The reasons for separating or
integrating the teaching of the four skills in the U. S. and the U. K. are reviewed in turn and in
their historical contexts.
In the early 1940s and during World War II, a group of specialists under the auspices of
the Linguistic Society of America were called on to develop effective, efficient, and intensive
language teaching to members of the U. S. Armed Forces. Based on methods for linguistic field
studies, and in keeping with the outline developed by Bloomfield (1942), the program worked
with a wide range of languages, such as Chinese and Hungarian, and was designed for target
language instruction in small classes of specially selected learners who were highly motivated.
The students were tutored by native-speaking informants together with linguists whose task was
to interpret the structural, lexical, and phonetic patterns of the language for teaching purposes.
The learners then drilled the elicited systematic patterns of spoken language to replace their first-
language "habits" with second language behaviors (Mitchell & Vidal, 2001).
In this way, following the principles of structural linguistics in conjunction with the
prevailing behaviorist learning theories, the primacy of speaking skills was established in a
famously successful language-teaching program. The instruction in, and the learning of, spoken
patterns was accompanied by similar structure-based teaching and learning of listening skills,
needed for conversing in a target language. Learning to read -- or write -- in another language
was not a focus of the linguistic analyses or of teaching simply because these skills were not
expected of the learners in their practical and required language uses in the field.
In a parallel development and in conjunction with teaching English as a second language
in the U. S., Charles Fries and his successor, Robert Lado, undertook to design a similar program
one solidly rooted in structural linguistics -- designed to teach English as a second or foreign
language. The implementation of the English language courses at the University of Michigan led
to the creation of the first North American set of teaching materials incorporating the English
sound system, common grammatical structures, and lexical patterns, excerpted from the available
linguistic analyses (Fries, 1945; 1952). In accordance with the principles of structural
linguistics, linguistic knowledge was methodically arranged for instruction in the first North
American course of its kind. The structural separation of second language skills and the primacy
of speaking served as a model for course and materials development. In later years, Robert Lado
(1957; 1964) formalized methods for contrastive and structural analyses of languages and their
application to the teaching and testing of discrete language skills. Lado's study (1957) was the
first systematic application of contrastive analysis to curriculum development, preparing
teaching materials and the discrete testing of such incremental language skills as phonemic
discrimination, vocabulary, and grammar.
The structural and behaviorist approach to language teaching and learning, with an
almost exclusive focus on speaking and grammar drills and listening comprehension, became
known as the oral method, the aural-oral method, the structural method, and in the 1950s
as the audiolingual method. Ellis (1990, p. 21) comments, however, that "audiolingualism was
very much an American method. In its purist form it was never very popular in Britain and
Europe, where less attention was paid to teaching the formal patterns of the second language and
more to their situational uses." Ellis continues, however, that "many of the audiolingual
assumptions regarding the way language is learnt can be found in pedagogical prescriptions of
British and European methodologists writing at this time."
In the U. K., the separation of the four skills had to do with the utilitarian purposes of
language teaching, rather than with matters of a particular methodology. According to Howatt
and Widdowson (2004), in the U.K., the period immediately after World War II was
characterized by "continuation and consolidation rather than change" in English language
teaching (ELT). Generally speaking, prior to the late 1950s, much of the British work in second
or foreign language pedagogy was devoted to teaching English to school-age children in the
colonies. Thus, teaching efforts were largely directed toward learners outside the U.K. without
much prominence attached to listening and speaking skills, but with a primary instructional focus
on learning grammar needed for translating written texts. The teaching of English as a second--
rather than a foreign--language was conducted primarily in London, and second language
teaching and learning there required listening and speaking skills, essential in basic
communication and routine interactions.
In the 1960s, however, with the influx of foreign workers and students, as well as former
colonials, British perspectives on ELT began to change. One of the top priorities in teaching
English as both a second and foreign language lay in the need for specialized instruction for
technical and highly trained personnel, for academic linguistic skills for college students, as well
as for grade school teachers of the children of immigrants. The emerging learning needs of these
new populations of learners brought about curricular and methodological work in two novel
directions: English for specific purposes for technical and professional learners and English for
academic purposes for university students. Thus, the language learning needs of specific groups
of learners led to pivotal shifts in the types of language that were taught, but not necessarily in
the specific skills that were taught. Howatt and Widdowson (2004: 247) comment that in the
1960s "history intervened in a somewhat dramatic way in the U. K., creating a wholly new
professional alignment." New and urgent demands arose for teaching the language needed in
technical and academic fields, as well as for integrated teaching of the discrete skills.
To this end, in the U. K., the emphasis on situational, rather than structural, language
skills became predominant in the curricula, similar to the syllabuses developed by A. S. Hornby
in the 1950s. The situational approach (also called the situational-structural method, the
structural-situational approach, or situational language leaching) resembled the pragmatic --
and situationalized -- version of the audiolingual method, with a primary emphasis on speaking
and listening skills. These were, however, socially, rather than structurally driven. According to
Howatt and Widdowson (2004: 299-300), the principles underlying language instruction between
the 1950s and 1970s postulated, "all four language skills (listening, speaking, reading, and
writing) should be taught, but the spoken skills should be given priority."
Real-world situational contexts of instruction, with lessons built around specific topics,
such as "at the post office," "at the doctor's," or "a visit to the theater," served as a backdrop for
teaching language "chunks" and contextually relevant grammar and vocabulary. The emergence
of the situational approach also gave rise to the classroom teaching technique currently known
as PPP (Presentation, Practice, Production). The PPP model of instruction implies that learners
can be guided from controlled practice of language features to free and automatic production of
language in any or all of the four skills. By the end of the 1960s, however, many linguists and
ELT methodologists arrived at the conclusion that the situational method was somewhat limited
in scope and its interactional foci -- i.e., speaking and listening -- and did not provide clear
principles that could guide curricula and instruction (Strevens, 1977).
While Stern (1983: 167) called the 1960s "disorienting" in language teaching and new
theories about language, the 1970s and 1980s ushered in humanistic approaches to language
pedagogy in North America and in Europe. It should be noted, however, that in the U. K. the
strongly pragmatic goal-orientation in teaching the four skills and the impact of social factors on
language usage continued to occupy a prominent place in ELT. (See, e.g., the work of Stevick
and Widdowson published in the 1970s for thorough discussions.)
The introduction of the concept of "communicative competence" (Hymes, 1971; 1972)
brought about a change in the perspectives on how language skills were to be taught and used for
communication inside and outside of the classroom. Although not directly associated with
language teaching per se, Hymes' work emphasized the key role of the social context in
communication and the centrality of the socio-linguistic norms of appropriateness in speech
communities and their cultures. Hymes was particularly interested in language as social
behavior. New perspectives began to emerge that authentic representations and uses of language
in the classroom were nearly impossible -- particularly so within the established models
associated with the audiolingual method. The structural separation of the four skills, pattern
practice, error avoidance, and native-speaker imitation in second and foreign language
production contrasted markedly with teaching language as a means of communication.
Communicative Language Teaching (CLT) places a great deal of value on
teaching language skills with the goal of enabling learners to communicate meaningfully
both inside and outside of the classroom, as in, for example, asking for information,
seeking clarification, relying on circumlocution when necessary, and in general,
negotiating meaning by all linguistic and non-linguistic means at one's disposal. In their
seminal publication on learners' coping strategies, Canale and Swain (1980) developed a
three-component framework of language competence that learners needed to achieve:
communicative competence, grammatical competence, and sociolinguistic competence.
Canale's and Swain's empirical findings demonstrated convincingly that practicing a
range of language skills simultaneously and in the context of communication allowed
learners to attain levels of grammatical competence similar to those achieved by students
who concentrated on audiolingual structural patterns. In addition, however, the
communicative competence of the learners who practiced their skills in interaction,
measured in terms of language fluency, comprehensibility, and effort, substantively
exceeded that of learners without comparable practice. As an outcome of this and other
studies published at the time (e.g., Paulston, 1974; Savignon 1972; 1983), CLT and its
subsequent methodological offshoots have presently come to dominate integrated
approaches to teaching of the central four skills. (See also Chapters 22 on Content-based
language teaching and 23 on Task-based language teaching.)

Linguistic and Methodological Bases for Integrating the Four Skills
As early as the 1970s, many researchers and methodologists noted that the teaching of
language skills can not be conducted through isolable and discrete structural elements (Corder,
1971; 1978; Kaplan, 1970; Stern, 1992). In reality, it is rare for language skills to be used in
isolation; e.g., both speaking and listening comprehension are needed in a conversation and, in
some contexts, reading or listening and making notes is likely to be almost as common as having
a conversation. The central innovative characteristic of the communicative approach in second
or foreign language teaching was the integration of the four macro-skills and their components.
Widdowson (1978) was one of the first linguists to call for integrating the four language
skills in instruction to raise learners' proficiency levels and enable advanced language learning.
In his proposal for integrated and communicative language teaching in general and in particular
in English for specific purposes, Widdowson emphasized that virtually all language uses take
place in the form of discourse and in specific social contexts. Although he notes that the
separated teaching of language skills is probably more administratively convenient, as in "divide
and rule" (1978: 144), language comprehension and production does not in fact take place in
discrete "units." Thus, to attain proficiency, learners need to develop receptive and productive
skills in both spoken and written discourse. Widdowson's (1978) strong emphasis on the
integration of the four skills, as well as discourse-based teaching, have had a considerable impact
on the emergence of discourse-oriented curricula and teaching methods in English for specific
purposes and English for academic purposes. Widdowson's (1978) and Halliday's (1978) early
work and their insights into the importance of discourse in language usage provided highly
influential theoretical foundations in linguistic analyses and language teaching. These works
have led to the subsequent rise and prominence of content-based and integrated language
instruction, especially in English as a second language in Australia, in the U. K., and, to some
extent in North America.
In the 1980s and 1990s, a great deal of elaboration and refinement took place in
communicative and integrated teaching of the four skills. In light of the fact that
opportunities for meaningful communication in the language classroom are limited --
particularly so in the regions where English is taught as a foreign language -- a great
need arose for integrated communicative activities. These had to be interaction-centered
and as authentic as possible to enable students to use the language for purposeful
communication (Savignon, 1983; 1990). The need for integrated activities led to the
evolution of task-based instruction that gained currency in the early to mid-1980s. (See
Chapters 22 and 23 for additional discussion.) At present, the ubiquitous language
practice exercises for groups or pairs of learners typically combine listening and
speaking, reading and speaking, or reading, writing, and speaking. Such integrated
classroom activities (also called tasks), include, for example, listening to language tapes,
playing games, or working on information gap and problem-solving exercises. These
types of practice require learners to engage in interaction and integrated language usage
because group- or pair-work can be carried out only if the participants share and discuss,
or read and pool their information. Task-based teaching is probably the most widely
adopted model of integrated language teaching today, and it is often considered to be the
closest classroom simulation of real-life interaction.
In his highly acclaimed book, Nunan (1989) outlines the principles that should guide the
design of teaching materials and modules for integrating a variety of language skills. (See also
Willis, 1996, for another set of such principles.) According to Nunan, effective integrated
modules are characterized by uses of authentic language models and exemplars, continuity of
language work from comprehension to production, explicit connections of classroom language
practice to real world uses (e.g., a business presentation or a job interview), and a systematic
language focus that enables learners to identify and analyze language regularities. In his later
work on designing integrated syllabuses, Nunan (2001) explains that the first step is to identify
the contexts and situations in which learners will need to communicate. After the
communicative events are identified in general terms, the next phase should work toward
learners' functional goals along with the linguistic elements required to achieve them. According
to Nunan, in integrated instruction, language skills are taught and practiced depending on the
students' learning objectives, rather than in the context of the four separate instructional areas.

Key Considerations in Integrated Language Teaching and Curricula
Richards, Platt, and Weber (1988: 144) define the teaching of integrated skills in
the Longman Dictionary of Applied Linguistics: "the teaching of the language skills of
reading, writing, listening, and speaking in conjunction with each other as when a lesson
involves activities that relate listening and speaking to reading and writing." There are
several principled models for integrating the teaching of two or more language skills.
Such models can vary substantially in their complexity and in the types of skills that can
be integrated to benefit learning, and virtually all have their advantages and
disadvantages in particular contexts.
The simplest and most basic type of integrated teaching incorporates the skills in
the same language medium, either spoken to include listening and speaking, or written to
include reading and writing. A typical instructional paradigm found in many locations
around the world deals with employing learners' receptive skills to provide input and
modeling for productive skills. For instance, in the spoken medium, listening selections
are used as models for speaking, interaction, or pronunciation skills, and in the written
medium, reading input supplies models for writing.
More complex integrated curricula combine a range of language skills. For
instance, instructional activities can bring together listening and reading input to promote
speaking or writing, or to facilitate both speaking and writing. In complex integrated
teaching methods, such as text-based (also called genre-based) language input
materials are usually organized thematically. In this way, theme, register, and language
content can be made consistent and cohesive to expose learners' to contextually linked
vocabulary, relevant grammar constructions, and discourse organization features. For
example, if the theme of the instructional materials has a focus on, say, weather, climate,
or geography, then in speech or writing, the register is likely to be somewhat more formal
than it would be in a module on friends and family. The vocabulary on weather and
geography is bound to include common climate-related terms, and the grammar
constructions are likely to deal with the present (but not the past) tenses, several kinds of
adverb clauses, and locational prepositions.
In practically all methods and techniques geared toward integrated teaching,
curricula typically include at least two essential teaching and learning objectives:
language features needed for communication and used in the context of
communication;
thematic and cohesive stretches of discourse for language input, rather than a
focus
on discrete vocabulary items, patterns, or grammar points.
Discourse-based approaches to instruction afford learners an opportunity to focus
on the linguistic and sociocultural features of organizing and presenting information in
particular contexts. (See Kaplan 1997; 2001; 2005 for a more thorough analysis.)
Teaching the language with a discourse focus also greatly facilitates an integration of a
broad range of skills when incremental skills can be transferred from one aspect of
language to another. For example, learning to organize and explain one's ideas in writing
can prove to be highly useful in structuring oral presentations. Similarly, the language
features -- e.g., vocabulary and grammar, associated with the formal register in speech --
can also be applicable to constructing semi-formal written text, such as an email to a
colleague. McCarthy (2001: 54) comments that transferability of skills from one type of
discourse to another provides for "a greater integration of the traditional four skills in
language teaching, where writing tasks might be 'spoken' in their mode, and vice
versa."
It should be noted, however, that the teaching of integrated language skills can
also have a number of disadvantages (e.g., McDonough and Shaw 2003; Widdowson,
1978, 1993; 2003). To begin with, a curriculum that concentrates on a single language
skill at a time can permit more focused teaching and more intensive learning.
Furthermore, in various regions and cultures where the instruction in discrete language
skills is highly valued, both teachers and learners have been known to resist skill
integration (Richards & Rodgers, 2001). In such settings, integrated instruction may not
be well suited to the local traditions of how teaching and learning are to be conducted.
Additionally, complex integrated instruction with more than two language skills
addressed in tandem places greater demands on both the teacher and the learner.
Curricula and syllabi that integrate a range of language skills require the teacher to be
reasonably versatile and well-trained. In most cases, the teachers need to be at least
somewhat familiar with discourse-based instructional models, such as those noted earlier.
At the same time, teachers can be expected to devote more time and effort to preparing
materials appropriate for integrated instruction. In many regions around the world, where
teachers are required to teach very large classes, the teaching of integrated skills may not
be a very practical option.
Another notable disadvantage of integrated instruction is that many (if not most)
learners have unevenly developed proficiencies across the four macro-skills (Hinkel,
2002; 2003; Stern, 1983). For example, second language learners who live in English-
speaking countries may have stronger skills in listening and speaking than in reading and
writing. Conversely, English as a foreign language learners are likely to be better readers
and writers than listeners and speakers. For this reason, the teaching of integrated skills
can become complicated, when instructional materials and practice have to account for a
considerable variance in learners' abilities. In complex integrated teaching, a frequent
tendency is for a particular language skill or set of skills to receive less attention than
learners' proficiencies might require. In light of the fact that the integrated curricula
concentrate primarily on purposeful communication and meaning making, typically, the
teaching of grammar and vocabulary, as well as accuracy in learner language production,
may receive less emphasis than they should (Richards, 2005). Some experts and
methodological authorities also contend that integrated language teaching with its main
focus on the learning process tends to overlook the quality of the learning product (e.g.,
Swan, 2005; Widdowson, 1990; 2003).

Language Tests and Testing of the Four Skills
It seems reasonable to expect that the methodological currents in language testing would
follow those prevalent in language teaching. However, the evolution of major standardized tests
has not coincided with the shifts and developments in teaching. In the 1960s, the newly
emerging standardized tests of English as a foreign language were constructed in keeping with
the prevailing influence of structural linguistics and the audiolingual method. At that time,
standardized English language tests were designed to assess learners' proficiencies in the skills
considered to be important for the overall mastery of the language. These incremental skills
included sound discrimination and listening comprehension, conversational and idiomatic
expressions, grammatical and structural knowledge, and vocabulary needed for reading
comprehension. In line with the standards in psychometric measurements of the time, to ensure
their reliability and validity, language tests consisted of large numbers of discrete-point multiple-
choice items.
Generally speaking, the design of discrete-point tests (also called discrete item tests)
assumes that language proficiency and its measurements encompass an array of components,
such as speaking, listening, reading, and writing and their smaller increments -- e.g., the sound
system, vocabulary, clause- and phrase-level grammar, or morphology (word forms and word
grammar) (see, e.g., Bachman, 1990; Bachman & Palmer, 1996; Brindley, 2001; for earlier
discussions, see J ones & Spolsky, 1975; Harris, 1969; Lado, 1961; NEC, 1962; Oller & Perkins,
1978; Valette, 1977). These incremental units of language -- i.e., the discrete points -- have been
the backbone of language testing for the more than a century in practically all regions of the
world.
In the U. K., the shift in the prevailing methodology was prominently reflected in the
design of language tests. The British English language examinations, such as that of the
University of Cambridge Local Examinations Syndicate (UCLES), were revised to make them
communicative and integrated. However, these tests were famously unreliable and error-prone.
According to Spolsky (1995), for example:
written compositions were marked subjectively,
listening comprehension was scored by clerical personnel according to a key, and
the speaking segments of the examinations were rated locally wherever they were
administered across several dozens of examination sites around the world.
In fact, Spolsky refers to three sources of measurement error:
i. a lack of consistency in the discrete-point listening and reading comprehension
sections with consistency coefficients so low as to make them statistically invalid,
ii. subjective marking of compositions without a means of establishing rater
consistency, and
iii. a high degree of variation in different forms of the test and of examiners' scoring
methods.
To some degree, the complex integrative and communicative nature of the examinations, the
complexities of their interconnected design, and the administration and scoring procedures also
caused a substantial amount of variation in test scores. As a result, UCLES examinations scores
were frequently seen as markedly unreliable and inconsistent (Bachman & Palmer, 1990). After
several studies and attempts to make Cambridge examinations more consistent and equivalent
across the forms and ratings, the integrative connections between the reading and the speaking
segments were removed in 1989, and between the reading and the writing modules in 1995.
According to the examination flyer, "both changes were widely welcomed by teachers and
students."
Spolsky (1995, p. 341) notes sardonically, "UCLES managed for a long time to resist the
claims of objective testing," just as the Educational Testing Service that constructs, administers,
and scores language tests in North America "managed to hold off the claims of integrative
communicative testing for some decades." Test design in North America did not adapt to the
shift from discrete-point testing of vocabulary, grammar, and listening comprehension until
2005, and even then, the integration of vocabulary testing went only as far as being incorporated
in the reading comprehension section. As an outcome, since the 1970s, important philosophical
and practical disparities have remained between the objectives of integrated language teaching
and the methods of language testing; that is, if the goal of communicative language teaching is
to enable learners to communicate meaningfully and appropriately in various contexts then,
clearly, discrete-point tests are inadequate and unsuitable for measuring learners' communication
abilities. For example, Paulston, and Bruder (1975) argued that, in real-life interaction, there is
always more than one way of effectively accomplishing a speaker's communicative goals. On
the other hand, even in carefully designed test items, there is only one correct answer, and such
an approach to testing that seems to resemble "the cue-response pattern" (1975: 15) conflicts
with the very purpose of teaching language for communication.
When communicative language teaching became the gold standard, the disconnect
between the foci of integrated instruction and the means of testing language proficiency as well
as the mastery of communicative skills gave rise to strong demands for similarly integrative tests
and testing. It was at that time that innovative types of language tests were created for the
purposes of institutional or local assessment; e.g., dictation tests that integrate listening and
writing abilities, or cloze tests that integrate vocabulary, grammar, and discourse skills (see, for
example, Oller, 1979; Oller & Perkins, 1978; 1980). In the 1970s and 1980s, a number of
integrative testing models were proposed for indirect measurements of learners' overall general
proficiency and communicative competence. However, after a series of experimental studies,
such test designs did not prove to be valid in different contexts for a variety of reasons.
For instance, research on proficiency measurements in various regions of the world --
e.g., Brazil and the Philippines -- demonstrated convincingly that language proficiency was not a
unitary trait, but that proficiency consists of complex and multi-faceted linguistic and
communicative components (e.g., Cohen, 1994; Farhady, 1982). Research in students' language
performance found that the mastery of one language skill does not necessarily translate to
competencies in any other skills. Furthermore, a great deal of evidence was obtained to show
that overall language proficiency cannot be adequately assessed by means of indirect tests; e.g., a
successful performance on an integrative vocabulary and grammar test has little to say about a
learner's writing abilities. Consequently, the separation of the four skills and the reliance on
testing discrete skills has been considered essential for accurate measurements of learners'
language proficiency -- or subsets of proficiencies. For example, an individual can have
excellent listening comprehension skills but a comparatively low speaking ability, and, as most
teachers know from experience, having had a great deal of grammar practice does not necessarily
lead learners to writing well.
Language tests and assessment instruments have remained largely unaltered in their
separation of the four skills since at least the 1960s. In the 1990s and early 2000s, a few
modifications have been made to provide at least some degree of integration in writing,
grammar, and vocabulary in North American standardized tests. For instance, overt and direct
testing of discrete grammar points has been eliminated, and assessments of grammar mastery are
now embedded in the essay tests, similar to the incorporation of vocabulary testing in reading
comprehension tests. By and large, however, language testing methods and test designs have
continued to rely on the separation of the four skills and skill subsets probably because a better
model for measuring language abilities has not yet been discovered. A fundamental issue in
language testing is that the skills needed for communication and communicative competence
seem to be enormous in their range and scope. There is little doubt, however, that the dichotomy
between communicative -- and integrated -- language teaching and discrete-point language
testing has had an indelible effect on the continued separation of the four skills in pedagogy,
research, curricular models, and teacher training (e.g., Hinkel, 2004; Hughes, 2003; Stern, 1992).
One of the greatest ironies associated with integrated teaching is that language proficiency
testing -- and hence much of the administrative gate-keeping -- has remained discrete-point-
based. Not surprisingly, though, not all learners set out to learn another language in order to do
well on tests.

Current Perspectives on Integrated Teaching
With the spread of English as a lingua franca and as the medium for world-wide
dissemination of information and knowledge, in many cases, the pragmatic objectives of
language learning underscore the importance of integrated and flexible instruction. In
many regions around the world, learning English has the objective of enabling learners to
gain access to social, vocational, educational, or professional opportunities (Celce-
Murcia, 2001; Kaplan, 1986; 1988; 1991). In common perspectives on contemporary
language curricula, teaching reading is typically connected to instruction on writing and
vocabulary, teaching writing can be easily tied to reading and grammar, and speaking
skills readily lend themselves to teaching listening, pronunciation, and cross-cultural
pragmatics (Hinkel, 1999; 2001).
According to Richards and Rodgers (2001: 165), integrated language instruction
that engages learners in meaningful communication and enables them to attain their
learning objectives can be found in an "unlimited" array of models, teaching materials
and techniques. A few examples of such integrated models with a communicative and
contextualized focus are: content-based (sometimes also called theme-based), task-based,
text-based (also called genre-based), discourse-based, project-based, network-based,
technology-based, corpus-based, interaction-based, literature-based, literacy-based,
community-based, competency-based, or standards-based.
With the current emphasis on both fluency and accuracy in language production,
it seems clear, however, that integrated language teaching and learning, as well as
integrative instructional models, will need to continue to be refined and developed (e.g.,
Breen, 1991; Swain, 1991). For instance, exposure to and experience with L2 speaking
and meaningful interaction, but without the benefit of explicit and focused instruction,
leads to learners' developing high degrees of fluency but not necessarily of accuracy and
advanced L2 proficiency (e.g., Lightbown & Spada, 1990; Swain, 1991).
In regard to communication-oriented principles that guide much of integrated
language teaching, critics have contended that with its focus on communication-in-
interaction, second or foreign language instruction frequently lacks depth and substance.
As Howatt and Widdowson (2004) note, naturalistic and integrated language learning
tends to meet the communicative needs that people would have as tourists in, for
example, simple service transactions and casual conversational exchanges. Widdowson
(2003: 24) explains, for instance, that "coping with written language is also a
communicative objective," but the contemporary focus on face-to-face encounters lacks a
teaching focus on "an understanding of writing, literary and otherwise, of the past." He
also points out that, outside English-speaking areas where English is taught as a school
subject, current methodologists would do well to consider "what kind of language is to be
specified for the subject to fulfill its educational objective" (2003: 27).
On the other hand, in recent years, standards- and outcomes-based language
teaching curricula have become one of the foremost educational priorities in a number of
English-speaking nations. To this end, innovative integrated methodological models
have been proposed to concentrate on advancing learner proficiencies in a range of
language skills. Specifically, the objectives of these models are geared toward clearly
defined language competencies that students need to achieve within the educational
system. Stern (1992) is to be credited with the first set of guidelines for an integrated
curriculum that address the major goal of advancing students' language proficiency. His
model effectively combines the learning of the central language skills with the
achievement-oriented syllabusi in culture learning, communicative skills, and general
education. Stern also notes (1992: 76) that "as useful expressions of proficiency,
however, the 'four skills' continue to be important categories in language pedagogy."
In Canada, for example, the instructional model is based on Stern's "proficiency
as competence" (1992: 73) and on communicative competence together with "the mastery
of such skills as listening, speaking, reading and writing." Canadian Language
Benchmarks and common sets of proficiency standards have been a recognized success in
nation-wide second language teaching. Canadian language assessments and national
achievement standards also account for the fact that many learners' language
proficiencies vary from skill to skill. In part, the effectiveness of the Canadian
achievement-oriented curriculum can be attributed to its design supporting learners with
different levels of mastery in the four skills (Breen, 2001). Other types of integrated
syllabusi are currently adopted in national educational movements and in the U. S.,
Australia, and New Zealand. These standards-based curricula reflect an ongoing work in
the refinement and elaboration of integrated instruction that can raise learners' language
proficiency and the quality of production in the contexts of real-life communication. (See,
e.g., McDonough & Shaw, 2003 for detailed discussion.)

Conclusion
In the past several decades, much evidence has emerged that, in order for learners
to attain language competence, teaching needs to integrate linguistic and communicative
skills. The overarching goal of integrated instruction is to advance learners' language
proficiency required for communication in various contexts. In general, the learning of
language for communication in both speaking and writing entails achieving mastery in
discourse, language strategies, sociocultural and interactional norms, and the
communicative culture of the people who use the language (Stern, 1992). Today, after
decades of research in language teaching and learning, it seems clear that, in many cases
and for many purposes, the separation of the four macro skills is likely to be less effective
than integrated instruction simply because, in reality, communication does not take place
in terms of discrete linguistic skills.
The early models of integrated and communicative teaching largely eschewed
explicit instruction in any of the four skills, and in particular in grammar. Typically, the
teaching of reading was integrated with writing, and listening with speaking. Classroom
instruction concentrated predominantly on activities and interactions -- e.g., on games,
role-plays, skits, and problem-solving in groups or pairs of learners. These exercises
sought to promote authentic language usage with the goal of developing learners' fluency.
The overarching objective of integrated and thematic language input was to facilitate
language acquisition naturalistically. At the time, the purpose of engaging with language
in classroom interactions was to enable learners to attain communicative competence. By
and large, learners' engagement in integrated communicative activities without deliberate
teaching led to incidental learning of such various linguistic features as conversational
expressions, vocabulary associated with daily and routine interactions, and informal
reading and writing skills (Hinkel, 2006).
In the mid-1990s, a number of studies found that many years of exposure to and
immersion in integrated classroom instruction does in fact lead to the development of
language fluency. It also became clear, however, that learners' productive language
lacked accuracy and sociocultural appropriateness, i.e., advanced communicative
competence that the integration of language skills in teaching sought to achieve (e.g.,
Lightbown & Spada, 1990; Swain, 1991).
The current models of integrated teaching of the four language skills have the
objective of developing learners fluency and accuracy, as well as their sociocultural
communicative competence requiring adapting the language from context to context and
from genre to genre. In light of the fact that at the present time English is widely
employed as the medium of international communication, it seems easy to predict that
integrated language teaching will continue to dominate among the various types of
pedagogical models. There is little doubt, however, that the evolution and change of
integrated teaching models and methods will remain one of the main -- if not the main --
defining characteristic of language teaching around the world.